# prepare data
num_samples = 500
X, y = datasets.make_moons(n_samples=num_samples,
noise=0.05,
random_state=0)
# shift label from {0, 1} to {-1, 1}
y = y * 2 - np.ones(len(y))
x_train, x_val, y_train, y_test = train_test_split(X,
y,
test_size=0.25,
random_state=40)
# build dataset
dataset = qnn.get_dataset_from_numpy(x_train,
y_train,
val_features=x_val,
val_labels=y_test)
# customize preprocessing
preprocessing_params = {"polynomial_degree": 2}
# customize model
model_params = {
"architecture": "qubit",
"num_q": 5,
"encoding": "angle",
"use_early_stopper": True,
"early_stopper_patience": 20,
"max_iterations": 10000,
"interface": "autograd",
"layer_type": "template",
# prepare data
train_features = np.linspace(-1. * np.pi, 1. * np.pi, 50)
train_labels = np.asarray((1 + np.sin(train_features)) / 2)
train_features = train_features / np.pi
val_features = np.linspace(-1. * np.pi, 1. * np.pi, 50)
val_labels = np.asarray((1 + np.sin(val_features)) / 2)
val_features = val_features / np.pi
train_features = train_features.reshape((len(train_features), 1))
val_features = val_features.reshape((len(val_features), 1))
plt.plot(train_features, train_labels)
plt.savefig("sinusoid.png")
dataset = qnn.get_dataset_from_numpy(train_features,
train_labels,
val_features=val_features,
val_labels=val_labels)
# customize postprocessing
model_params = {
"architecture": "cv",
"num_q": 2,
"encoding": "displacement",
"snapshot_frequency": 5,
"prefix": "sinusoid",
"early_stopper_patience": 50,
"use_early_stopper": True,
"num_layers": 5,
"interface": "autograd",
"layer_type": "template",
}
from sklearn import datasets
from sklearn.model_selection import train_test_split
import prevision_quantum_nn as qnn
if __name__ == "__main__":
# prepare data
X, y = datasets.make_moons(n_samples=200, noise=0.05, random_state=0)
x_train, x_val, y_train, y_test = train_test_split(X,
y,
test_size=0.25,
random_state=42)
dataset = qnn.get_dataset_from_numpy(x_train,
y_train,
x_val=x_val,
y_val=y_test)
application = qnn.get_application("classification")
application.solve(dataset)